Aliso: A Journal of Systematic and Evolutionary Botany Volume 10 | Issue 4 Article 11 1984 Redescription and Distribution of Muilla Coronata (Liliaceae) James R. Shevock Follow this and additional works at: http://scholarship.claremont.edu/aliso Part of the Botany Commons Recommended Citation Shevock, James R. (1984) "Redescription and Distribution of Muilla Coronata (Liliaceae)," Aliso: A Journal of Systematic and Evolutionary Botany: Vol. 10: Iss. 4, Article 11. Available at: http://scholarship.claremont.edu/aliso/vol10/iss4/11 ALISO 10(4), 1984, pp. 621-627 REDESCRIPTION AND DISTRIBUTION OF MUILLA CORONATA (LILIACEAE) James R. Shevock INTRODUCTION Muilla coronata Greene was first collected by the botanical pioneer C. C. Parry in March 1888, Mojave Desert (near Landcaster), Los Angeles County, California. Described by Edward Greene (1888) in the same year, this rel­ atively rare species was known only from the type specimen for 34 years. During a May 1983 collecting trip on the Kern Plateau, southern Sierra Nevada of Tulare County, the author discovered a population of several thousand Muilla coronata plants. At that time it was assumed that these plants would prove to be a new subspecies. This assumption was based on two primary morphological characters: the presence of only one leaf per plant and perianth segments nearly twice as long as described. This new material, however, matched perfectly with plants labelled as Muilla cora nata in California herbaria. One possible explanation for these inconsistencies could be that Greene was compelled to work from a single fragmentary specimen (i.e., corms, leaves, flowers not attached). His reconstruction of the specimen would have been partially based on Sereno Watson's description of the genus Muilla. Other species in the genus were known to have several leaves. However, the type specimen (ND) contains four plants, two with corms attached to the scape which contain but one leaf. An alternative explanation could be that the type was a different taxon than the specimens attributed as Muilla co­ ronata in California herbaria. However, the material in the herbaria labelled as M. coronata resemble the type specimen. Another possible explanation is that Greene thought that some of the leaves were detached and lost during the collection of the Parry specimen. Evidence of detached leaves from the corm would be nearly impossible to detect or refute. The author speculates that this may have been the source of the error. Greene's description, though extremely accurate in the discussion of the greatly dilated petaloid filaments, assumed other characters that are not evident from the Parry collection. These erroneous features have been per­ petuated in both monographic and floristic works (Abrams 1923, Ingram 1953, Munz 1959 and 1974). Greene (1888) makes reference to Watson's well-proposed genus Muilla while describing M. coronata. The other two species known at that time [M. maritima (Torr.) Watson and M. transmon­ tana Greene] both have several leaves arising from a fibro-membranous 622 ALISO Fig. I. Close-up of Muilla coronata flowers displaying the crown of petaloid filaments. The diminutive size of this Wy is readily apparent when compared to the quarter used for scale which is 23 mm in diameter. Scape is less than 3 em above the sandy granitic gravels and the plant has only one leaf. (Bar, upper right = 5 mm.) coated corm. Though the type of M. coronata (ND) shows no indication that the leaves are "several," Greene nonetheless states "leaves 2-3" in the original description. Review of herbarium material also strongly contradicts the leaves 2-3 statement. Only one plant was observed with two leaves (Mason 14259, UC) with the second leaf being rudimentary. Field observations of several thousand plants from the Kern Plateau failed to find even one plant with two leaves. Though Ingram (1953) describes 2-3 leaves per plant, the three cited collections of M. coronata in his monograph contain plants with a single leaf. Abrams (1923) also described M. coronata with leaves 2-3 in the description, but the illustration (Abrams 1923:398) reveals only one leaf. In addition, the illustration displays the strongly dilated petaloid filaments radiating outward against the perianth segments. However, in nature the filaments are erect though wholly distinct from one another forming a cy­ lindrical crown reminiscent of several Trite/eia species (Fig. 1). Other characteristics in the original description that require clarification are the number of flowers per umbel, length and color ofperianth segments, VOLUME I 0, NUMBER 4 623 and the length of the scape. Greene ( 18 8 8) states that M. coronala is 3-10 flowered. The type specimen contains four plants, three of these with four flowers per umbel and the remaining one with six flowers. The majority of M. coronata examined have 2-5 flowers, though rarely, a robust individual will contain up to 8(-11) flowers. The perianth segments are described as 1.5-2 lines long (approx. 3.2-4.3 mm) though the dried specimens and live material studied range from 5 to 9 mm. Greene further states the color of the perianth from light blue or nearly white within and exteriorly green with bluish margins. All plants observed by the author are white with a broad green line on the exterior of the perianth segments which darkens nearly black on dried specimens. Corms can be 2.5-8 em below the surface de­ pending on soil type and composition. The scape of M. coronata above ground is very short. compared to other Muilla species. Generally, in M. coronata, the scape is less than 3 em, with robust plants to 5 em above the ground. The leaf is 2-2.5 times as long as the scape. On dried specimens, there is a distinct band marking on the scape showing that portion which occurred above and below the ground surface. REDISCRIPTION OF MUILLA CORONATA Corm 1-2 em thick, 2.5-8 em below ground surface; scapes very slender, less than 3(-5) em above ground; leaves one (rarely two), narrowly linear, semiterete, 2-2.5 times as long as the scape, the margins retrorsely scabrous; umbels 2-4 bracted with 2-5 rarely 8(-11) flowers; perianth rotate, the segments 5-9 mm long, white with a green line on the exterior; filaments greatly dilated, hyaline-petaloid, cuneate oblong in outline, obtuse, retuse or almost obcordate above, forming a cylindrical crown; anthers subsagittate, erect, fixed by the middle or a little above it to an incurved median acu­ mination of the broad filament. Capsules globose, 3-angled, loculicidal; seeds black. KEY TO THE SPECIES OF MUILLA Perianth white, never yellow Leaves several, equaling or slightly longer than the scape; fila­ ments filiform, subulate or dilated at base Filaments filiform to subulate, not dilated at base; anthers generally purplish (though occasionally pale yellow) M. maritima Filaments dilated at base forming a cuplike corona; anthers yellow M. transmontana Leaves one (rarely two) 2-2.5 times as long as the scape; filaments petaloid, widely dilated, forming a cylindrical crown; anthers yellow M. coronata Perianth yellow, never white M. clevelandii 624 ALISO Fig. 2-3.-2. Typical habitat for Muilla coronata in a pinyon pine woodland on the Kern Plateau, Tulare County. Populations occur on the gentle sloping ridgetops in the foreground with the Dome Land Wilderness, Sequoia National Forest in the distance.-3. More detailed VOLUME 10, NUMBER 4 625 DISTRIBUTION The distribution of Muilla coronata is primarily restricted to the Mojave Desert in a shadscale scrub and Joshua tree woodland (Munz 1959). Botan­ ical field work in the Owens Valley and adjacent areas by Mary DeDecker has located populations of Muilla coronata in a creosote bush and sagebrush scrub community. Recent collections in the New York Mts. (Thorne 43338, RSA), Scodie Mts. (Shevock 10796, CAS), and the southern Kern Plateau (Shevock 10454, CAS, MO, ND, NY, RSA, SBBG; Shevock 10809, CAS, MO, RSA; and Shevock 10824, CAS, JEPS, LAM, RSA) extend the habitat into a pinyon-juniper woodland community (Fig. 2-3). The upper portion ofthe population on the Kern Plateau contains an occasional weather-beaten Jeffrey pine. Populations are known for Inyo, Kern, Los Angeles, San Ber­ nardino, and Tulare counties, California, eastward to Clark County, Nevada, ranging from 701 to 2012 m in elevation. Representative specimens: CALIFORNIA. Inyo County: Devils Kitchen S of Coso Rd., Zemba! 520 (RSA); 1 mi W of Independence, Kerr 488 (CAS); 1.5 miN oflndependence, Munz s.n. (Lenz 1975); SE ofMt. Whitney Fish Hatchery, DeDecker 84 (pers. herb.); Kern County: Iron Canyon, El Paso Mts., Weston s.n. (CAS); near Muroc Dry Lake, Heckard & Moe 4539 (JEPS); Indian Wells Valley, DeDecker 796 (pers. herb.); 2 mi W of Lamont Peak, Kern Plateau, Shevock 10809 (CAS, MO, RSA); 4.5 mi SW of Long Valley, Dome Land Wilderness, Sequoia Nat'l. Forest, Shevock 10824, (CAS, JEPS, LAM, RSA); 1 mi NE of Pinyon Peak, Scodie Mts., Sequoia Nat'l. Forest, Shevock 10796 (CAS); Los Angeles County: Mojave Desert, Parry s.n. (ND) and Sherwood s.n. (LAM); Ave. J & 170 St. near Joshua Tree State Park, Wilsons. n. (Lenz 1975, and pers. comm.); San Bernardino Coun­ ty: Mouth of Deep Creek, Hesperia Lake Arrowhead Rd., Jaeger s.n. (CAS); N of Kramer, Mason 11727 (UC) and Mason 14259 (UC); Caruthers Can­ yon, New York Mts., Thorne 43338 (RSA); Tulare County: 7.2 mi W of Lamont Mdw. on rd. to Long Valley, Kern Plateau, Shevock 10417 (CAS, MO, ND, NY, RSA, SBBG) and Shevock 10454 (CAS, MO, ND, NY, RSA, SBBG). NEVADA. Clark County: Red Rock Canyon, Spring Range, Baci­ galupi & Ferris 6263 (JEPS). Other Muilla coronata populations discovered by Mary DeDecker where collections were not obtained include: Inyo County, N of Fort Independence, 4 mi N of Independence Courthouse; about Owens Lake, 0.5 mi N of Cot­ tonwood Charcoal kilns; and SE of Carricut Lake, Naval Weapons Center.